From: Namhyung Kim <namhyung@kernel.org>
To: Ian Rogers <irogers@google.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>,
Jiri Olsa <jolsa@kernel.org>,
Adrian Hunter <adrian.hunter@intel.com>,
Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@kernel.org>,
LKML <linux-kernel@vger.kernel.org>,
linux-perf-users@vger.kernel.org
Subject: Re: [PATCH 2/4] perf hist: Simplify hist printing logic for group events
Date: Wed, 14 Feb 2024 21:18:00 -0800 [thread overview]
Message-ID: <CAM9d7cinDt0W73KJPaMBgWkG0oe3Kds7RC75pzqMJo+kfDkuTw@mail.gmail.com> (raw)
In-Reply-To: <CAP-5=fWnVFabJ8MqsnAQAC5Bbk9AQmcnTCWJNrd3RGk8Ucoe-w@mail.gmail.com>
On Wed, Feb 14, 2024 at 3:57 PM Ian Rogers <irogers@google.com> wrote:
>
> On Mon, Feb 12, 2024 at 11:53 PM Namhyung Kim <namhyung@kernel.org> wrote:
> >
> > It can uses an array to save the period (or percent) values for member
> > events and print them together in a loop. This simplify the logic to
> > handle missing samples in members with zero values.
> >
> > No functional change intended.
> >
> > Signed-off-by: Namhyung Kim <namhyung@kernel.org>
> > ---
> > tools/perf/ui/hist.c | 55 +++++++++++++++++---------------------------
> > 1 file changed, 21 insertions(+), 34 deletions(-)
> >
> > diff --git a/tools/perf/ui/hist.c b/tools/perf/ui/hist.c
> > index 2bf959d08354..5f4c110d840f 100644
> > --- a/tools/perf/ui/hist.c
> > +++ b/tools/perf/ui/hist.c
> > @@ -33,6 +33,7 @@ static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he,
> > char *buf = hpp->buf;
> > size_t size = hpp->size;
> >
> > + /* print stand-alone or group leader events separately */
> > if (fmt_percent) {
> > double percent = 0.0;
> > u64 total = hists__total_period(hists);
> > @@ -45,12 +46,19 @@ static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he,
> > ret = hpp__call_print_fn(hpp, print_fn, fmt, len, get_field(he));
> >
> > if (evsel__is_group_event(evsel)) {
> > - int prev_idx, idx_delta;
> > + int idx;
> > struct hist_entry *pair;
> > int nr_members = evsel->core.nr_members;
> > + union {
> > + u64 period;
> > + double percent;
> > + } *val;
> >
> > - prev_idx = evsel__group_idx(evsel);
> > + val = calloc(nr_members, sizeof(*val));
> > + if (val == NULL)
> > + return 0;
>
> Looks good, but should this return value be negative to indicate an
> error? Snprintf gives a negative result on error, its result is
> sometimes returned from hpp__fmt_acc, as is this result.
Sounds good, will change.
Thanks,
Namhyung
>
> >
> > + /* collect values in the group members */
> > list_for_each_entry(pair, &he->pairs.head, pairs.node) {
> > u64 period = get_field(pair);
> > u64 total = hists__total_period(pair->hists);
> > @@ -59,47 +67,26 @@ static int __hpp__fmt(struct perf_hpp *hpp, struct hist_entry *he,
> > continue;
> >
> > evsel = hists_to_evsel(pair->hists);
> > - idx_delta = evsel__group_idx(evsel) - prev_idx - 1;
> > -
> > - while (idx_delta--) {
> > - /*
> > - * zero-fill group members in the middle which
> > - * have no sample
> > - */
> > - if (fmt_percent) {
> > - ret += hpp__call_print_fn(hpp, print_fn,
> > - fmt, len, 0.0);
> > - } else {
> > - ret += hpp__call_print_fn(hpp, print_fn,
> > - fmt, len, 0ULL);
> > - }
> > - }
> > -
> > - if (fmt_percent) {
> > - ret += hpp__call_print_fn(hpp, print_fn, fmt, len,
> > - 100.0 * period / total);
> > - } else {
> > - ret += hpp__call_print_fn(hpp, print_fn, fmt,
> > - len, period);
> > - }
> > + idx = evsel__group_idx(evsel);
> >
> > - prev_idx = evsel__group_idx(evsel);
> > + if (fmt_percent)
> > + val[idx].percent = 100.0 * period / total;
> > + else
> > + val[idx].period = period;
> > }
> >
> > - idx_delta = nr_members - prev_idx - 1;
> > -
> > - while (idx_delta--) {
> > - /*
> > - * zero-fill group members at last which have no sample
> > - */
> > + /* idx starts from 1 to skip the leader event */
> > + for (idx = 1; idx < nr_members; idx++) {
> > if (fmt_percent) {
> > ret += hpp__call_print_fn(hpp, print_fn,
> > - fmt, len, 0.0);
> > + fmt, len, val[idx].percent);
> > } else {
> > ret += hpp__call_print_fn(hpp, print_fn,
> > - fmt, len, 0ULL);
> > + fmt, len, val[idx].period);
> > }
> > }
> > +
> > + free(val);
> > }
> >
> > /*
> > --
> > 2.43.0.687.g38aa6559b0-goog
> >
next prev parent reply other threads:[~2024-02-15 5:18 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-13 7:52 [PATCHSET 0/4] perf report: Omit dummy events in the output (v1) Namhyung Kim
2024-02-13 7:52 ` [PATCH 1/4] libperf evlist: Update group info in perf_evlist__remove() Namhyung Kim
2024-02-14 23:50 ` Ian Rogers
2024-02-15 5:16 ` Namhyung Kim
2024-02-13 7:52 ` [PATCH 2/4] perf hist: Simplify hist printing logic for group events Namhyung Kim
2024-02-14 23:57 ` Ian Rogers
2024-02-15 5:18 ` Namhyung Kim [this message]
2024-02-13 7:52 ` [PATCH 3/4] perf hist: Do not use event index in hpp__fmt() Namhyung Kim
2024-02-15 0:08 ` Ian Rogers
2024-02-15 5:26 ` Namhyung Kim
2024-02-15 7:54 ` Ian Rogers
2024-02-16 20:08 ` Namhyung Kim
2024-02-13 7:52 ` [PATCH 4/4] perf report: Do not show dummy events with --skip-empty Namhyung Kim
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAM9d7cinDt0W73KJPaMBgWkG0oe3Kds7RC75pzqMJo+kfDkuTw@mail.gmail.com \
--to=namhyung@kernel.org \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=irogers@google.com \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).